EFFECT OF CONSTRUCTION METHOD ON SHEAR RESISTANCE OF CONCRETE MASONRY WALLS

Transcription

1 EFFECT OF CONSTRUCTION METHOD ON SHEAR RESISTANCE OF CONCRETE MASONRY WALLS Oan, Ahmed Faisal 1 ; Shrive, Nigel 2 1 PhD, Candidate, University of Calgary, Civil Engineering Department, afoan@ucalgary.ca 2 Professor, University of Calgary, Civil Engineering Department, ngshrive@ucalgary.ca Bed joint reinforcement is used in concrete masonry walls to improve resistance against both in- and out-of-plane lateral loading. There is no doubt that the use of joint reinforcement enhances the post-cracking performance of masonry walls, but there are divergent results in the literature as to whether the use of joint reinforcement increases or decreases the overall shear resistance of masonry walls. It could be that the method of construction is the cause of the discrepancy in results, so an experimental program was executed to study the effect of two different ways of placing the bed joint reinforcement in the walls. The first method was to lay the bed joint reinforcement on the dry blocks with the mortar subsequently being placed on top, whereas the second method was to place the mortar on the face-shells and then embed the joint reinforcement in the mortar. The first method is the method used on site in Canada, while the second method would supposedly give better results as there should be no unbonded area under the reinforcement. Tests were carried out on 12 walls (1.6 m long by 1.4 m high) under in-plane vertical and lateral loading. The two methods of construction were each applied to three partially grouted and three ungrouted walls. Statistical analysis of the results with the T-test showed that there was no difference in the shear strength obtained between the two ways of laying the bed joint reinforcement in the walls. Keywords: Construction methods, masonry, shear resistance. INTRODUCTION Bed joint shear reinforcement has been shown to increase the shear resistance and improve the post crack behaviour of concrete masonry walls when placed in conjunction with vertical steel [Scrivener (1969), Shing et al. (1989),Khattab and Drysdale (1993), Voon and Ingham (2002)]. However, when bed joint shear reinforcement is utilized on its own [e.g. Hatzinikolas et al. (1980), Foltz and Yancey (1993), Schultz et al. (2000), Gouveia and Lourenco (2007), Jasinski (2010)], differences can be noted with respect to the effect on the overall shear resistance of the walls. For example, Gouveia and Lourenco (2007) showed slight increase (5%-10%) in shear resistance when using bed joint reinforcement in unreinforced masonry, whereas Hatzinikolas et al. (1980) observed a significant decrease in the ultimate compressive strength of their masonry walls when bed joint reinforcement was included in the construction. Oan and Shrive (2009) also showed a reduction in the shear strength when applying bed joint reinforcement to the unreinforced masonry walls. In all cases, there is a consistent improvement in post-cracking behaviour.

2 Therefore, due to the inconsistency in results with respect to the effect of bed joint reinforcement, an experimental program was executed to determine if the difference in results was due to the method of placing the joint reinforcement in the concrete masonry walls. EXPERIMENTAL PROGRAM Twelve concrete masonry walls, each 1590 mm long (four blocks) and 1390 mm high (seven courses), were tested under bi-axial monotonic loading. The walls were divided into two equal groups: ungrouted and partially grouted walls. In each group, three replicates of each of two different ways of placing the bed joint reinforcement were constructed. The two different ways of incorporating the bed joint reinforcement in the bed joints are shown in Figure 1. Groups of three walls were used as the minimum to allow elementary statistical analysis to check the significance of the results obtained. The masonry materials were also characterized, namely the concrete masonry units, the mortar, the grout, the joint reinforcement and the masonry assemblage. MATERIAL PROPERTIES Hollow concrete masonry units of nominal dimensions 400*200*200 mm and average compressive strength of 23.2 MPa (five specimens) were used in the construction of the walls. All the walls were constructed in face-shell bedding by the same experienced mason with type S mortar (CSA 2004). Six 50 mm mortar cubes were taken from each batch of mortar and tested for compressive strength. Six 100*200 mm cylinders were taken from each grout mix, with three being tested at an age of 7 days and the remaining three being tested at 28 days to obtain the compressive strength of the grout. The compressive strength of mortar cubes at the age of 28 days varied from 12.8 MPa to 15.3 MPa (24 mortar cubes were tested) while the compressive strength of the 12 grout cylinders tested varied from 18.2 MPa to 24.5 MPa. The average compressive strengths of the mortar cubes and grout cylinders for the different walls are listed in Table 1 along with the standard deviation, where three specimens were tested at each age (7days and 28 days) for both mortar cubes and grout cylinders. Ten 3-high prisms were built at same time as the walls. The average compressive strength of these prisms was 15 MPa 28 days after construction based on the net areas of the block and grout. The mortar joints in the walls and prisms were tooled to obtain concave mortar joints - this tooling increased the density of the mortar at the edge of the joints. Ladder type bed joint reinforcement with 3.7 mm diameter steel wires was used. The average yield stress (3 specimens were tested) for this joint reinforcement was 530 MPa.

3 (a) (b) Figure 1. Methods of laying bed joint reinforcement: (a) On Dry surface, (b) Embedded in mortar

4 TEST ARRANGEMENT AND INSTRUMENTATION The test arrangement consisted of two vertical hydraulic actuators to apply axial load and a horizontally mounted actuator to apply lateral load to the top two courses of the wall, as shown in Figure 2. The vertical actuators each had a maximum load capacity of 1.5 MN and a maximum stroke of 250 mm, while the horizontal actuator had a maximum load capacity of 500 kn and maximum stroke of 150 mm. To distribute the vertical load over the length of the wall, a steel I-beam was placed on top of the wall, with a layer of fiberboard between the stiff distributor beam and the wall. The walls were built on a steel C-channel which was fixed to the floor by bolts to prevent in- and out-of-plane displacements. Four steel dowels 180 mm high were welded along the steel beam, and the cores in the masonry into which these dowels protruded were either fully grouted to the top of the wall for the partially grouted walls or grouted only in the first course for ungrouted walls. This was necessary to prevent the uplift of the walls under high lateral loads. As each wall had eight cores, for the partially grouted walls cores 1, 3, 6 and 8 were grouted. Five displacement transducers were installed along the height of the walls on both sides to monitor the lateral displacement and another two transducers were installed at the lower course to measure any in-plane rotation of the walls, as shown in Figure 2. Figure2. Test setup and instrumentation.

5 TESTING PROCEDURE The static bi-axial test was performed on two stages. Axial load was first applied to the walls with the vertical actuators at a rate of approximately 1 kn/s until the required level of axial stress was reached. The required axial load was calculated based on the net effective crosssectional area of the walls. The desired axial stress was 3 MPa. The axial actuators were then placed in force control and programmed to maintain the same axial load through the remainder of the test. In the second stage, shear load was applied via the horizontal actuator at an imposed displacement rate of 0.1 mm/s in displacement control. The test was stopped when the lateral load had dropped to 80% of the maximum value achieved. Data on the loads and displacements were collected on a PC using the LABTECH data acquisition program. STATISTICAL METHOD Results were compared through a T- test to determine whether there were significant differences between the two methods of incorporating the bed joint reinforcement in the walls. EXPERIMENTAL RESULTS AND DISCUSSION All the walls failed in a mixed flexural and shear mode where cracks first appeared in the bed joints in the tension side of the walls followed by the formation of diagonal cracking and crushing at the toe at the compression side of the wall. Typical cracking is shown in Figure 3. Individual load results are presented in Table 2. Figure 3. Typical failure mode for the tested walls. The lateral load was applied to the top left corner of this wall.

6 Table1. Compressive Strength of mortar and grout for different walls. Mortar Grout Wall 7-days 28- days 7-days 28- days # f mortar SD* f mortar SD* f grout SD* f grout SD* Ungrouted Embedded Ungrouted On Dry surface Grouted Embedded Grouted On Dry surface *SD=standard deviation Table2. Properties of different tested walls. Wall # Cracking Load (kn) Max Load (kn) Max. Displacement (mm) Energy Dissipation (kn.mm) Ungrouted Embedded Ungrouted On Dry surface Grouted Embedded Grouted On Dry surface

7 UNGROUTED WALLS For the ungrouted walls, the average maximum shear resistance when laying the bed joint reinforced embedded in mortar was kn which increased to an average of kn when the joint reinforcement was laid directly on dry face shell of the blocks (Table 3). The 2.7 % difference is negligible. Indeed, the T-test results showed that statistically there was no significant difference between the two values, and thus no effect on maximum load between the two ways of applying the reinforcement as shown in Table 3. In order to compare the ductility of the walls, the area under the load-displacement curve was calculated to measure the energy dissipation. This energy was found to be 2540 kn.mm on average for the embedded reinforcement walls and was 2550 kn.mm for the walls with reinforcement laid on dry surface, again almost the same for the two cases as shown in Table 4. PARTIALLY GROUTED WALLS For the partially grouted walls; laying the bed joint reinforced embedded in mortar resulted in an average shear resistance of kn, while laying the joint reinforced on dry surface resulted in an average shear resistance of kn. This 2.6% difference is statistically insignificant when the T-test is applied to the data. Results of the T-test for the partially grouted walls are shown in Table 3. When comparing the energy dissipation between the two ways of applying the reinforcement in case of partially grouted walls, embedding the joint reinforcement in the mortar resulted in an average energy dissipation of 3470 kn.mm. This value appears to be higher than that obtained from laying the joint reinforcement on the dry surface (3130 kn.mm). However, the result of analyzing the data with the T-test showed that the difference is statistically not significant as shown in Table 4. There is a greater difference in the energy dissipation (10%) between the two forms of placing the reinforcement in case of partially grouted walls compared to the difference in the ungrouted walls (0.4%). EFFECT OF GROUTING The average shear resistance of the walls with embedded reinforcement increased from kn to kn with grouting of the walls. In these tests therefore, grouting provided a 65% increase in the resistance of the walls although they had the same level of axial stress (3 MPa). Also, in case of placing the joint reinforcement on the dry surface and then placing the mortar, the shear resistance increased from kn to kn with grouting - a 57% increase. In both cases therefore, there was a substantial increase in shear resistance through grouting. However, if the strength is considered on a stress basis, that is we consider the applied shear load resisted by the net area, the shear strength of the ungrouted walls was higher than that of the partially grouted walls. For the ungrouted walls, for both embedded shear reinforcement and for the reinforcement laid on the dry block before the mortar is applied, the average shear resistance of the walls was 2 MPa. For the partially grouted walls, embedding the reinforcement gave an average strength of 1.7 MPa while laying on the dry block gave an average of 1.6 MPa. The implication is that there is a 15% to 20% reduction in shear strength due to grouting.

8 Table 3. T-test results for shear resistance (peak lateral load). Ungrouted Ungrouted Partially Grouted Embedded Rft. Rft. on Dry surface Embedded Rft Partially Grouted Rft. on Dry surface Mean Difference between Means Standard Deviation COV P-Value Result Statistically not significantly different Statistically not significantly different Table 4. T-test results for energy dissipation (area under the load-displacement curve up to the point where the post-peak load is 80% of the peak). Ungrouted Ungrouted Partially Grouted Partially Grouted Embedded Rft. Rft. on Dry surface Embedded Rft Rft. on Dry surface Mean Difference between Means Standard Deviation COV P-Value Result Statistically not significantly different Statistically not significantly different CONCLUSIONS For the walls tested in this study, the two methods of placing the bed joint reinforcement showed no difference in the overall shear resistance of the walls. In addition, neither of the two ways of construction affected the initiation of cracks in the walls nor the energy dissipation in the walls. These observations were confirmed with T-test comparisons. Grouting the concrete masonry walls increases the load capacity but decreases the shear strength of the walls when considering the net area resisting the applied loads. ACKNOWLEDGEMENTS This work was performed with the financial support of the Natural Sciences and Engineering Research Council of Canada, and the Canadian Concrete Masonry Producers Association. The help of the technical staff of the Department of Civil Engineering, the University of Calgary and the support of the Masonry Contractors Association of Alberta (South) is gratefully acknowledged

9 REFERENCES CSA S Design of Masonry Structures Canadian Standards Association, Mississauga, Ontario,2004. Foltz Stuart and Yancey, W.C. Influence of Horizontal Reinforcement on the Shear Performance of Concrete Masonry Walls", ASTM Special Technical Publication, No 1180, 1993, pp Gouveia, J. P. and Lourenco, P. B. Masonry Shear Walls Subjected to Cyclic Loading: Influence of Confinement and Horizontal Reinforcement,10th North American Masonry Conference, June 3-6, Missouri, USA, 2007, pp Hatzinikolas, M., Longworth, J. and Warwaruk, J. Failure Modes for Eccentrically Loaded Concrete Block Masonry Walls. Journal of the Am. Conc. Institute, Vol77, 4, 1980, pp Jasinski Radoslaw. Researches of Reinforced Clay Brick Masonry Walls Horizontally Sheared. 8th International, July 4-7. Dresden, Germany, Khattab, M.M. and Drysdale, R. G. The Effect of Reinforcement on the Shear Response of Grouted Concrete Masonry. The Masonry Society Journal, Vol. 12, No.1, 1993, pp Oan, A. F. and Shrive, N. G. Shear of Concrete Masonry Walls. 11 th Canadian Masonry Symposium, May 31- June 3,Toronto, Ontario, Canada,2009. Schultz, A. E., Hutchinson, R.S. and Cheok, G.C. Seismic Performance of Masonry Walls With Bed Joint Reinforcement. 12 th International Brick/Block, June 25-28, Madrid, Spain, 2000, pp Scrivener, J.C. Static Racking Tests on Concrete Masonry Walls. Proceedings of the International Conference on Masonry Structural Systems,, University of Texas, 1969, pp Shing, P.B., Noland, J.L., Klamerus,E. and Spaeh.H. Inelastic Behavior of Concrete Masonry Walls. Journal of Structural Engineering, Vol. 15, No.9, September, 1989, pp Voon, K.C. and Ingham, J.M. Experimental Study of the Shear Strength of Reinforced Concrete Masonry Walls. 6th International, 4-6 Nov, London, United Kingdom, 2002, pp